Cognitive Spatial Representative Map for Interactive Conversational Model of Service Robot

• Published in: 2018 27th IEEE International Symposium on Robot and Human Interactive Communication (RO-MAN) pp. 686 - 691
• Main Authors: Bandara, H. M. Ravindu T., Muthugala, M. A. Viraj J., Jayasekara, A. G. Buddhika P., Chandima, D. P.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2018
• Subjects: cognitive map; conversation model; Data mining; Human robot interaction; Navigation; Service robots; Shape; social robotics; Spatial databases
• ISSN: 1944-9437
• DOI: 10.1109/ROMAN.2018.8525778

Assistive robots are developed for supporting the daily activities of human beings to uplift the living standards. Assistive robots should be friendly, reliable, active, comprehensible and capable of creating interactive conversations with users in order to be a friendly companion for the human. Humans tend to include uncertain terms related to directions and distances to describe or express ideas. Furthermore, an assistive robot should be capable of analyzing and understanding the numerical meaning of uncertain terms for the purpose of creating a cognitive map in order to build friendly interactions between the user and the robot. Moreover, this paper proposes a method to identify the spatial relation between the objects in a given environment and describes such environments using uncertain terms related to spatial information based on a cognitive map created by the proposed system while having interactive conversations with the user. Conversation Management Module (CMM) and Spatial Information Processor (ISP) and Cognitive Map Creator (CMC) have been introduced in order to create interactive conversations while processing the uncertain information based on a cognitive map. Capabilities of the robot has been demonstrated and validated from the experimental result.